Data Analytics
Self-Service BI
Data Cleaning
Browser-native
Plotly.js
Zero-install
DataLens BI
DataLens BI is a browser-based analytics workspace for CSV and Excel files. It helps users inspect data quality, clean fields, build visuals and explain patterns without sending files to a server.
100%
Browser — zero upload
⚠ The problem
Non-technical teams receive raw CSVs with no way to assess quality, apply cleaning, or explore patterns — creating analyst bottlenecks that delay decisions by days or weeks.
✓ The solution
Upload → auto-profile → one-click clean → interactive dashboards with plain-English insights. Everything runs in the browser. Data never leaves the device.
🔒
Privacy-first architecture. All data processing happens in the visitor's own browser tab using JavaScript.
No file is ever transmitted to any server. Sessions are fully isolated and reset on close.
Where this applies
NGO / M&E Teams
Field Data Quality Reviews
M&E officers upload monthly field collection exports and instantly see quality scores, missing indicators, and outlier flags before donor submission.
Finance Teams
Budget Variance Exploration
Finance analysts upload expenditure data to generate spend-by-category KPIs, trend charts, and exception flags — without waiting for the BI team.
HR / Operations
People Data Self-Service
HR managers explore headcount, attrition, and tenure distributions on HRIS exports — without sharing sensitive data with any third-party tools.
Analytics pipeline architecture
The platform is designed as a privacy-first browser pipeline: files are parsed locally, profiled into metadata, cleaned with an auditable transformation layer, then passed to reusable analytics and export modules.
1. Ingest
Local file loader
CSV and Excel files are read inside the browser using PapaParse and SheetJS. No source data is uploaded to a server.
2. Profile
Schema and quality scan
The engine detects data types, missing values, duplicates, outliers, date fields and usable measures for analysis.
3. Transform
Cleaning workspace
Auto-clean and manual operations standardise text, fill gaps, remove duplicates and cap outliers while keeping an audit trail.
4. Analyse
Reusable insight layer
Shared state powers KPIs, charts, pivots, trends, correlations, map views and plain-English insight summaries.
5. Publish
Export package
Users can download filtered data, cleaned data, pivot output, an HTML report or a multi-sheet Excel workbook.
Cross-cutting controls
Stability, privacy and usability guardrails
Browser-only processing
Selected-column state
Responsive chart containers
Filter-aware outputs
Cleaning audit log
Safe export reset
Platform capabilities
🎯
Auto Data Quality Score
Completeness, duplicate, and outlier scoring across all columns.
🧹
One-click Auto-Clean
Fills missing values, drops duplicates, caps outliers, standardises text — with full audit log.
💡
Dynamic Chart Insights
Plain-English finding generated below every chart — skew, peaks, % differences, correlations.
🔗
Correlation & Driver Analysis
Pearson heatmap, scatter with trend line, and driver ranking for any target metric.
📐
Pivot Table Builder
Row × column × value breakdowns with sum, avg, median, min, max aggregations.
🔒
Privacy-First Design
All processing in-browser. Nothing sent to any server. Safe for sensitive organisational data.
Technical stack
Parsing
PapaParse · SheetJS
Analytics engine
Vanilla JS ES2022 modules
Design
DM Sans · Portfolio theme
Deploy
Any static host — no server needed